In this article, Michele Del Mondo, Senior Director Global Advisor Automotive, PTC explains how software architecture is redefining the automotive industry.
Modular architecture, over-the-air updates and embedded artificial intelligence have become strategic cornerstones for automotive manufacturers. Among European OEMs, discussions around embedded software clearly illustrate the scale of the challenge ahead. Porsche, for instance, has publicly outlined the evolution of its software strategy for electric vehicles. At the same time, technology players such as Huawei are entering the mobility market with integrated platforms like HIMA, and even with their own brands.
Yet behind these initiatives, a fundamental divide is emerging. Beyond visible features, there are deeply different system architectures that underpin these two opposing approaches.
Established suppliers must deal with existing, often highly heterogeneous technical environments. By contrast, new entrants design their software architecture, data models and platform logics from a single source of truth. For the former, the challenge is one of profound transformation, for the latter, rapid scaling.
Many companies also carry decades of engineering legacy, and existing toolchains are simply not designed for software-defined vehicles (SDVs). They remain largely component-centric, while modern vehicles must now be primarily developed in a function-oriented manner.
From Components to System Functions
In the SDV, function becomes the primary unit of reference. A brake assist function, for example, is no longer just a control unit or a mechanical component, but a complete system combining software, sensors, hardware, safety requirements and approval certificates. Any change to this function therefore impacts several areas at once, including code, parts lists, test cases, variant rules, and regulatory documentation.
In reality, however, this information is still often scattered across disconnected systems. PLM, ALM, CAD and ERP coexist with limited interoperability. This fragmentation prevents a holistic view of the product and becomes particularly critical as the number of variants continues to explode: powertrains, software packages, local regulations and safety-related validation levels.
When Complexity Becomes Exponential
Today, an SDV is defined as much by its software requirements and configurations as by its hardware components. Every additional variant introduces yet another layer of complexity into embedded systems.
Without an integrated data fundation and systemic traceability, impact analysis remains a manual and time-consuming task. Answering seemingly simple questions like Which markets are affected? Which software version is installed in which vehicle? What test coverage applies to each configuration?, requires exhausting cross-team coordination.
The consequences are clear. Change cycles lengthen, validation risks increase, and time-to-market delays are driven less by technical hurdles than by organizational complexity resulting from fragmented data and tools.
This challenge extends far beyond the automotive industry. In many other sectors, variant management remains disconnected from core systems, causing field data to reach engineering teams too late. This disconnect has become a structural obstacle to competitiveness.
Towards an Intelligent Product Lifecycle
Regaining control requires a coherent data model across the entire product lifecycle. This is precisely the ambition of the Intelligent Product Lifecycle: to structurally connect all product attributes — from requirements and system models to part lists, software artefacts, test cases and configurations. As a result, the impact of change can be tracked in a systemic, auditable way.
Manufacturers and suppliers then share a single source of truth, allowing teams to shift their efforts away from manual data reconciliation and toward informed technical decision-making. The toolchain thus assumes a central connecting function within the value chain.
This consistency is essential to managing the growing number of variants. Even as functionalities continue to expand and update cycles accelerate, configuration traceability remains complete. Scaling becomes possible without losing control.
What Companies Need to Change
The focus is on harmonizing the data model. This means that functions, requirements, variants, and components must be defined in a semantically consistent manner and linked together systematically. Only once this structure is in place can ALM and PLM processes be integrated in a meaningful way. In the same way, configurations, versions and validations must not be managed separately by discipline but must be consistently traceable throughout the entire product lifecycle.
Changes must be embedded in model-driven product line engineering, structured through automated and auditable workflows that track both technical and regulatory impacts. In this context, toolchain consolidation goes far beyond IT optimization. It is a strategic architectural choice.
Whether transforming a complex industrial legacy, as established manufacturers must do, or building an SDV-native platform from the ground up, as new entrants can, one structural requirement remains constant: a coherent and integrated data and systems architecture in which functions, variants, and software versions are linked throughout.
In the long run, industrial competitiveness will not be determined by the number of tools deployed, but by the coherence of the overall ecosystem. Engineering has fundamentally become a discipline of data architecture.
